The Research And Application Of Network Video Processing Key Technology

With the popularity of the Internet,video business has showed its stronger influence,while the related video processing technology gets more and more attention naturally.How to make use of video processing technology to better meet the needs of video business,has always been a focus of research.Generally speaking,according to different requirements and application,during the process of storage,transmission and display playback process,the need to involve the relevant processing technology of video data is often not the same.For example,in the process of video files’ storage,transmission,the commonly used method is to adopt the way of container packaging for video file preprocessing.Although,the general approach is simple and reliable,there are some potential drawbacks,and even can’t meet the needs of some specific scenarios.Optimized for video preprocessing process naturally became urgent matters,and get the attention of people gradually.This paper makes a deep study of the video file coding,packaging technology,and summarizes some basic rules in the present commonly used video preprocessing process,and takes this as the breakthrough point,puts forward a new scheme to achieve the goal of video’s fast preprocessing.This paper also research the relevant copy processing technology of video data on the display playback process,proposes a corresponding optimization methods,through the analysis of existing performance defects which in the commonly used methods.This article’s main work is as follows:Firstly,this paper analyzes the basic principles of the existing video preprocessing techniques and shortcomings on its performance,put forward an improvement ideas to solve these defects,simplifies the existing video preprocessing process using the "hidden rules" which most videos follow,proposes the technique of "index type" video preprocessing,so as to achieve the original file wasn’t damaged and reducing the disk IO.Than analyzes and designs the framework of video preprocessing procedure on the whole,disposes the video of encapsulation,parsing,encoding and decoding function modularity in accordance with the format,realizes the decoupling between analytic and decoding operation of different video formats under the premise of unified process.Secondly,according to different types’ video,this paper gives details of the "index" methods,including indexed video preprocessing schemes of public video format(h.264,h.265,MPEG,etc.),and indexed video preprocessing schemes of private video formats(some vendors only provide parser,instead of segments ‘meaning openness).Thirdly,in order to handle the player’s performance problems result from processing a large number of video data in the player client,this paper utilizes the SSE multimedia instruction,combing with Microsoft DXVA hardware decoding output interface,realizes the technology copying large video frame from memory,thus improves the phenomenon of a high CPU occupancy rate resulting from the high resolution image copying.At the last,this paper designs and implements experiments and analyzes results to the two proposed means,then compares the traditional video preprocessing technology and indexed video preprocessing technology performance,traditional data copying technique and video hardware decoding data fast copying technique performance.Demonstrates the effectiveness of the suggested methods in this paper.